Welded Versus Solid Piece Modal Analysis

[duhuhu]

Looks like it is time for round two... So for any of you who were following my other post (https://www.physicsforums.com/showthread.php?t=762560) about whether or not to glue my model or make a solid piece for a modal analysis, you know that I was having a lot of trouble deciding. It turns out that the decision was made by my overseeing professor who said he wants a solid piece. Easy peasy right? WRONG! Well turns out that it is not that easy because my mentor now wants to weld the piece instead.

So now what I am looking for is some suggestions on whether or not to go with my mentors idea of welding it or to suggest that we do the solid milled piece. I know that there are some issues that come along with aluminum welding, though I am unsure of what they actually are, that could make the process a little difficult.

And to those of you just joining in, the problem is that I am creating a model out of aluminum that has unit cell dimensions of 6x6x6 cm for the main structure and then a pillar on top (non load bearing) that is 1.5x1.5x3 cm. these will be lined up in a long rod with the pillars on one side of the rod. the rod will be about 90 cm in length.

I am performing a Modal Analysis of this structure in the frequency range of 10-8000 Hz and need to know (now) if welding the structure would be a viable solution to our problem or if we should go with the solid piece.

Based off of the last discussion; I am still feeling solid piece, but it is not up to me. So I am looking for some solid reasoning, for or against welding or solid structure. Or if you have any other ideas that could be viable (unfortunately this does not include shrink fitting due to our shop moving)

Thanks!

 

[AlephZero]

A welded joint should behave like solid metal - except that if you are trying to weld a 1.5 x 1.5 x 3 cm "lump" onto a base, you probably have to choose between getting the weld to go right through the 1.5 x 1.5 section, or severely distorting the geometry of the welded part compared with an idealizaton of two rectangular blocks joined together.

Welding Al is perfectly possible if you do it right, but "buying a cheap welding kit meant for DIY car repairs and learning to use it by trial and error" won't work. The basic issues are that compared with steel, Al has high thermal conductivity, low melting point, and is very reactive once you take away the surface layer of oxide that coats all solid Al parts. But if your workshop has the facilities for inert gas welding and a technician who is a trained welder, no problem.

 

[duhuhu]

one thing that one of my buddies suggested was that if we were to weld around the outside, then the center would not actually be attached so it would act like an internal crack in the system which would severely distort our results. is there any way to work around this?

 

[AlephZero]

one thing that one of my buddies suggested was that if we were to weld around the outside, then the center would not actually be attached so it would act like an internal crack in the system which would severely distort our results.

I agree. That's why you would need the weld to penetrate all the way across the joint, but I don't see how you could do that and maintain the geometry accurately.

I suppose you could weld some over-sized lumps of metal onto the beam structure and them machine them to an accurate shape. That might be quicker and/or cheaper than machining the whole thing from solid.

Have you considered making the parts by casting? If you have the basic facilities for foundry work, you should be able to get pretty good results with simple technology like sand molding.

 

[mesa]

Looks like it is time for round two... So for any of you who were following my other post (https://www.physicsforums.com/showthread.php?t=762560) about whether or not to glue my model or make a solid piece for a modal analysis, you know that I was having a lot of trouble deciding. It turns out that the decision was made by my overseeing professor who said he wants a solid piece. Easy peasy right? WRONG! Well turns out that it is not that easy because my mentor now wants to weld the piece instead.

So now what I am looking for is some suggestions on whether or not to go with my mentors idea of welding it or to suggest that we do the solid milled piece. I know that there are some issues that come along with aluminum welding, though I am unsure of what they actually are, that could make the process a little difficult.

And to those of you just joining in, the problem is that I am creating a model out of aluminum that has unit cell dimensions of 6x6x6 cm for the main structure and then a pillar on top (non load bearing) that is 1.5x1.5x3 cm. these will be lined up in a long rod with the pillars on one side of the rod. the rod will be about 90 cm in length.

I am performing a Modal Analysis of this structure in the frequency range of 10-8000 Hz and need to know (now) if welding the structure would be a viable solution to our problem or if we should go with the solid piece.

Based off of the last discussion; I am still feeling solid piece, but it is not up to me. So I am looking for some solid reasoning, for or against welding or solid structure. Or if you have any other ideas that could be viable (unfortunately this does not include shrink fitting due to our shop moving)

Thanks!

It's not that big of a deal for someone experienced with welding however this project will require a good AC TIG (tungsten inert gas) welder which will be expensive. Miller and Lincoln make good machines.

The areas to be welded will require some prep (cutting grooves so there is good penetration while leaving a nice looking weld when finished). For someone inexperienced I would recommend going to the local community college welding department and hiring a student to do the work for you. If you decide to take the challenge on yourself then get some scrap and practice, practice, practice!

Another option is brazing. There are two routes you can go on this, the best way is with an oxy/acetylene torch set, the correct brazing rod, and lot's of practice. Option two is to get on Harbor Freight's website and order some Alumiweld 730 'brazing' rods plus a good propane or MAPP gas torch kit. With a little practice they can make a decent braze in a pinch.

 

[mesa]

I should also add, if you have a good Mill to use don't be afraid to try it out so long as you have someone to show you the ropes. They are wonderful machines!

 

[duhuhu]

Mesa, My office is actually directly above our universities best shop (The Aerospace Instrumentation Shop) so we have all the equipment to weld and mill here. We are busy expanding the shop though so we don't have access to shrink fitting, foundry, or any advanced fabrication methods while that is going on.

I appreciate all of the input that you guys have given, but milling is the easiest way to go it looks like, and it should give us good results, so that is the way to go. I will be sure to post if I need any more help.

 

[mesa]

That works, and it sounds like you won't be needing any of the equipment that is currently down. Good luck with the piece!